by Christian Scholz on September 25, 2009
Last week I was moderating a panel about open platforms at the Community Summit in Hamburg with member from german social networks studivz and Xing. One question discussed was of course Open Social as both networks either have an implementation running already or will shortly launch one.
The big difference in what an open platform might be summarized as follows:
- The users come to the applications (and take their data with them, think “distributed social graph”)
- The applications come to the user (as OpenSocial apps)
What is more open? What will win in the end?
Now both approaches have pros and cons of course but one of the disadvantages of the second approach is probably what I call “the App Store Filter Syndrome”. It basically means that the networks implementing OpenSocial are also the gatekeepers of what gets added as application and what doesn’t. It’s apparently quite similar of what the computer industry always dreamt of: Trusted Computing. It basically means that not the user decides what application she trusts but the hardware vendor or in this case the network vendor does.
And to me this is the big disadvantage of the application approach anyway. I am already missing useful apps on e.g. Xing and there is no way I can install them. I am dependant von Xing.
With studivz it will not look better as they also plan to approve applications before they get added.
No wonder that this wasn’t popular in the audience as it consists if many marketing professionals who also want to use OpenSocial as marketing means for their clients. They were very interested in the question how they could add applications to these networks without first implementing it and then not getting approved, thus losing money for development.
The future: A few big networks or many small ones?
It seems that with a few big networks the OpenSocial solution is the winner while with many small ones the distributed data model is the winner.
On the panel there wasn’t really consensus on which will win. Everybody still likes of course to lock in their users but some might be more realistic that this is not really what users want. A few big networks (lets say 2, one for business and one for personal uses) is more a scary thought and so I personally am really favouring a solution where users are free to choose which service the give their data to and that they are able to move easily between them.
What is your opinion?